A quick recap and then a few closing comments.
I wanted to end with a few comments on Knights in the Nightmare. As I explained in my review/repair of the game, the extremely cluttered design severely crippled the gameplay experience. The dense, front loaded tutorials delivered too much information the worst way possible: out of context. The cluttered visuals made understanding even the most basic gameplay states and interactions difficult. And the various stat based RPG systems for the characters and weapons are too complex for how simple they are to optimize and how shallow the gameplay is.
I consider most of what I played of Knights in the Nightmare's single player mode poorly designed; a clear example of design waste. After putting 15 hours into the game and playing 25 missions I found that the gameplay challenges worked against what little meaning the core had to convey. KitN is a shmup-strategy-RPG hybrid. Story considerations aside, this means that the core meaning is expressed through the these three types of gameplay systems. As I've explained this series, designing mandatory challenges that are different to a degree that is imperceptible or below the minimum degree of difference results in a poorly communicated interactive experience. Though I won't go into detail here, there are too many missions in KitN that force players to focus on minute changes in a game that's already far too cluttered. With the weapon and character RPG systems incredibly shallow for being so complex, the strategy gameplay lost in a cluttered real-time arena, and the shmup gameplay thrown on top of it all, it's clear that the core of KitN is difficult to convey cleanly. The levels are not designed around this fact.
Consider the fractal zoom video (above) as a perfect metaphor for the infinite undiscovery of video game design spaces and the art therein. As infinitely enthralling as this fractal is, there are still huge black areas and areas of flat, patterness color that would not be as interesting or meaningful to zoom in on. Consider the camera in the video as analogous to level design in that it guides the viewer/player along to focus on what's most meaningful. Also consider that because the video ends, there's a limit to how much one can zoom in on any particular spot; thus there's a limit to the complexity and meaning of the fractal video. That's the idea in a nutshell.
RPG systems inherently face functional clutter design issues because numerical values define the continuum of their continuous design spaces. The finest degree by which stats can be altered is usually not meaningful to the gameplay. And it's not just RPGs that we have to worry about. All games of decent complexity have a core they express and nuances to that core. There's also the potential to succinctly design levels to convey the core, and the potential to present too many levels that detract from the conveyance of the core. Because of emergence, as a game increases in complexity so too does potential negative side effects, which I call chaos.
With more chaos there are more emergent possibilities that are difficult to account for and control through design. Not all of these possibilities can be presently cleanly based on the feedback design of the game. Not all of these possibilities are something designers want the player to experience. Not all of these possibilities are coherent or make sense when presented to the player. Therefore as the complexity increase for game, so too does its non-meaningful possibilities. Compared to the coherent, patterns we find meaning in, there are infinite possibilities you don't want to discover in a complex design space. Yes, you can spend time and energy exploring these chaotic spaces with, but the returns diminish. This is why it's better to leave some potential of your design space untapped rather than to go overboard with redundant, excessive, distracting, or meaningless challenges.
Just remember, though I defend gameplay completely, I know that gameplay isn't everything. There are story, stylistic, musical, visual, and experiential reasons to bend and break the conventions of good game design. Though I think many appreciate this idea of "any thing goes" for art, the reality is, it takes a masterful artist to know exactly how far to bend and break the rules to achieve the desired effect without needlessly sacrificing the core principles and conventions of the medium.
In the end "infinite undiscovery" seems like such a simple concept, yet it took me 9 parts to explain it thoroughly. Most of the premises I constructed were things that many of us already knew intuitively. However, putting them together illuminated several ideas about the art of game design that aren't so intuitive. I could expound on this topic forever, but I think here is the best place to stop.